hs38DH reference file and HLA*NN:NN:NN sequence names

Question

[14134125465346445]

Is there a recommended way of pre-processing the hs28DH.fa file or post-processing the bam files that come out of the alignment process to deal with the funky naming convention for the HLA sequences?

Tools like Picard and deeptools don't deal well with sequence names like 'HLA*ABC1:01:02:03' .

Any recommendations?

[Devon Ryan]

Could you report the deepTools bug on github and provide me with an example that breaks it? We're heavily updating deepTools at the moment and it'd be good to get this fixed before the next release.